Migration, Microfinance and Mechanized Service Market:
Explaining the Puzzle of Landless Tenancy in Rural Bangladesh
Binayak Sen and Paul Dorosh
International Food Policy Research Institute (IFPRI)
Presentation made at the ReSAKSS-Asia Conference, 12-14 December 2017, Bangkok

Landless Tenancy and Upward Mobility:
Motivation for Research
• The previous literature—both Neo-Classical and Marxist--suggests that
agricultural tenancy is an inefficient form of contract and, with green revolution
as the farm operations become profitable, the importance of owner-operated
farming will rise and the share of land under share tenancy is likely to decline.
• The same literature also suggests that the group of landless tenants (the so-
called “pure tenant”) will become a vanishing tribe, as they become more
involved in non-farm and non-agricultural activities.
• None of these have happened in rural Bangladesh: the share of cultivated land
under tenancy has doubled and there has been a noticeable rise of landless
tenants. In addition, we see a distinct shift from share tenancy to fixed-rent
tenancy. We seek to explain this puzzle and show that this is related to the
nature of broader structural transformation in rural Bangladesh.
• To this end, we use HIES (2000 and 2010) for primary analysis supplemented by
the secondary evidence from the 62-Village Panel Surveys (1988, 2000 and 2008).
Explaining Landless Tenancy in Rural Bangladesh 2

The Rise of Landless Tenancy: Evidence from
the 62-Village Panel Survey (1)
• Share of Rented-In Land in Total Cultivated Land: In 1988, only 23.4%
of cultivated land was under tenancy in rural Bangladesh; the
matched share increased to 32.9% in 2000, rising further to 39.8% in
2008 (Hossain and Bayes 2009; Hossain, Sen and Sawada 2016).
• Share of Sharecroppers/ Leaseholders in Total Rural Households: In
1988, 43.6% of households rented land from others; the matched
share rose to 54.2% in 2000 and 58.3% in 2008.
Explaining Landless Tenancy in Rural Bangladesh 3

Rising Importance of Landless Tenancy: Evidence
from the MH Panel Survey of 62-Villages
0
10
20
30
40
50
60
70
Area under Tenancy (%) % of Tenant farms Area under Landless Tenancy (%) % of Landless Tenants
Expanding Tenancy Market and Rising Landless Tenancy
1988 2000 2008/ 2004
Explaining Landless Tenancy in Rural Bangladesh 4

The Rise of Landless Tenancy: Evidence from
the 62-Village Panel Survey (2)
• Share of Landless Tenants in Total Rural Households: In 1988, only
13.6 of rural households were landless tenants; the matched share
has increased to 20.5% in 2000, rising further to 27.6% in 2004
• Share of Landless Tenants in Total Cultivated Land: In 1988, only
6.7% of total cultivated land was under the landless tenancy; the
matched share has increased to 12.4% in 2000, rising further to 19.3%
in 2004
• This is also evidenced from HIES rounds for 2000 and 2010, which
form the basis of our quantitative results reported in this paper
Explaining Landless Tenancy in Rural Bangladesh 5

The Rise of Landless Tenancy: Evidence from
HIES (1)
Explaining Landless Tenancy in Rural Bangladesh 6
0
10
20
30
40
50
60
Landless Functional landless Marginal Small Medium Large
Distribution of Area under Tenancy by Land-Size (%)
2000 2010

The Rise of Landless Tenancy: Evidence from
HIES (2)
Explaining Landless Tenancy in Rural Bangladesh 7
0
10
20
30
40
50
60
Landless Functional landless Marginal Small Medium Large
Distribution of Tenant Farms by Land-Size (%)
2000 2010

Change from Sharecropping to Fixed Contract
• Forms of Tenancy Changed from Sharecropping to Fixed Rent:
According to 1960 Agricultural Census, 91% of land under tenancy
were cultivated under the sharecropping system. As per the 2008
Agricultural Census, 43% of land are cultivated under the
sharecropping system, the rest being cultivated under fixed rental
and/or long-term mortgage arrangements.
• Bangladesh is silently making transition from ‘inefficient’
sharecropping to ‘efficient’ fixed rental system. We argue that the
rise of landless tenancy, together with the decline in the share of
sharecropping as the form of rent payment, has had considerable
poverty reducing effects.
Explaining Landless Tenancy in Rural Bangladesh 8

Who Gives and Who Takes: Outlining an
Analytical Framework (1)
• Increased Supply of Farm Land to the Agricultural Tenancy Market:
• Supply of agricultural land to the tenancy market can increase due to rising
importance of non-agricultural incomes for agricultural landowners.
• Secondly, it may also increase due to rising ‘supervision cost’ in hired labor
based farming due to increased specialization in nonagricultural work.
• Thirdly, even those medium/ large farms who previously conducted family
labor based farming may opt for tenant based farming, facing demographic
decline in the supply of family labor.
• In the empirical work, we capture these tendencies by four variables:
salaried work (proxy for non-agricultural income), domestic migration and
foreign migration (proxy for supervision cost), and availability of male
worker (proxy for family labor).
Explaining Landless Tenancy in Rural Bangladesh 9

Who Gives and Who Takes: Outlining an
Analytical Framework (2)
• Increased Demand for Farm Land in the Agricultural Tenancy Market:
• Traditional theory suggests that demand for landless tenancy will be limited due
to shortage of financial capital and lack complementary non-land assets such as
draught power and irrigation pumps that are needed for cultivation.
• However, these traditional theories are based on ‘old’ production conditions
which no longer remain valid for rural Bangladesh.
• Thus, landless tenant households can address their financial capital need much
better than before through access to microfinance, which has expanded virtually
to each village of Bangladesh.
• Landless tenant households are no longer constrained by the lack of
complementary non-land inputs. There has been a rapid expansion of
mechanized service markets supporting agricultural operations (hiring power
tiller services substituting for animal draught power, buying irrigation water from
different private sources, and recently, taking recourse to mechanized thrashers).
Explaining Landless Tenancy in Rural Bangladesh 10

Who Rents in land: Results for ‘All Tenants’
and ‘Landless Tenants’ (1)
• What are the likelihood of being a tenant farmer as opposed to owner
farmer? To answer this question, we run a probit model for both HIES 2010
and HIES 2000 with adequate additional controls such as income level and
spatial fixed effects. Here we focus on the results for 2010.
• Access to mechanized service markets--access to irrigation and power
tiller services-- increases the probability of being in the tenant category
both for ‘all tenants’ and for ‘landless tenants’.
• Access to microfinance enhances the probability of being in the tenant
category, as it reduces the problem of raising finance. However, the result
is significant only for ‘all tenants’. This may suggest that microfinance
addresses the problem of start-up agricultural capital not for the pure
landless having no amount of land but for the other groups of the less poor
with some amount of land.
Explaining Landless Tenancy in Rural Bangladesh 11

Probability of Renting In vs. Owner Farming: Key Results for
2010 HIES
Variables All Tenants Landless tenants
Someone with salaried
jobs
-0.0420*** -0.0430***
Domestic remittance 0.0116 0.0247*
Foreign remittance -0.0185 -0.0168
Microcredit access 0.0182* 0.00637
Irrigation access 0.193*** 0.185***
Rented power tiller/
tractor
0.342*** 0.315***
HH Head’s education -0.00653*** -0.00388***
Female Headship -0.0893*** -0.0825***
Number of adult males 0.0232*** 0.0127*
HH Size 0.00660** 0.00322
Explaining Landless Tenancy in Rural Bangladesh 12

Who Rents in land: Results for ‘All Tenants’
and ‘Landless Tenants’ (2)
• Access to migration to cities acts as a stimulating factor for landless
tenants. Financial support through the route of domestic migration
helps them to pay for the cash rent.
• Availability of adult male workers adds to the pool of family labor
and thereby increases the chances of being in the tenant category.
• Access to salaried jobs (as well as human capital accumulation)
discourages tenant farming both for ‘all tenants’ and for ‘landless
tenants’. This finding may suggest the possibility of specialization for
the land-poorest—one opting for the route of non-agricultural jobs
and the remining other, specializing in agriculture.
Explaining Landless Tenancy in Rural Bangladesh 13

Who Rents out land: Results for ‘All Landlords’
and ‘Large Landlords’ (1)
• Renting out for common among the relatively large and medium sized
landowning groups. Thus, an active tenancy market contributes to
equitable land-distribution.
• We run the same probit model for exploring the decisions to rent out land.
Decisions for renting out seem to be associated with factors which are
almost the mirror images of factors correlated with decisions for renting in.
This is true for both the groups of ‘all landlords’ and ‘large landlords’.
• Thus, having salaried jobs and human capital encourages renting out.
Similarly, non-availability of male family labor encourages them to rent out
land. Female headship also motivates the households to rent out land.
Explaining Landless Tenancy in Rural Bangladesh 14

Probability of Renting Out vs. Owner Farming: Key Results for
2010 HIES
Variables All Landlords Large Landlords
Someone with salaried
jobs
0.0168* 0.0807**
Domestic remittance 0.0227** 0.0818*
Foreign remittance 0.0158 0.0531
Microcredit access 0.00826 0.0279
Irrigation access -0.0744*** -0.0783*
Rented power tiller/
tractor
-0.132*** -0.149***
HH Head’s education 0.00492*** 0.00775**
Female Headship 0.0303** 0.115**
Number of adult males -0.0231*** -0.0339
HH Size -0.00246 -0.00868
Explaining Landless Tenancy in Rural Bangladesh 15

Concluding Remarks
• The increase in the share of cultivable area under tenancy and the unpredictable rise of landless tenants
have been brought about by a confluence of several factors that underpinned rural structural
transformations in Bangladesh
• Spread of education, expansion of regular jobs, and urbanization (domestic migration) have encouraged
renting out of the agricultural lands on the part of land-rich households
• The growth of microfinance and the rapid development of market for mechanized services have made the
tenant farming on an increasing scale a feasible proposition. This is true in case of landless tenants as well.
• These transformations in the tenancy market are brought about not by decree (i.e. land tenure reform, as in
West Bengal) but by changing market conditions, and in that sense, they are market-led, not state-led
transformations
• There seems to be a trend of specialization going on among the landless households in rural Bangladesh:
some landless groups are entering into the land sector as tenants, while other landless groups are moving
out of the land sector, being increasingly involved in non-agricultural jobs.
• In conclusion, the agricultural route of upward mobility via landless tenancy must be recognized as equally
potent route as the non-farm route for uplifting the landless households out of poverty
Explaining Landless Tenancy in Rural Bangladesh 16